Virtual SOTIF Scene Testing for Autonomous Vehicle Failure Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current SOTIF testing methods for autonomous vehicles lack a unified testing method and specific risk evaluating standard, leading to inconsistent testing scenes and safety concerns for testers, with real road tests being inefficient and limited in scope.
Innovation Solution
A comprehensive SOTIF testing system and method that includes a scene database with scene groups corresponding to source types, a virtual testing module to generate and test virtual scenes, and an analyzing module to identify trigger conditions or function limitations causing failures, thereby providing improvement methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If real road tests are conducted for SOTIF testing, then the testing covers actual driving conditions, but the safety of testers is compromised and testing efficiency is low
Solution Approach 1:
The patent creates virtual copies of real road scenarios through a scene database containing various driving environments (rainy weather, nighttime, complex intersections). These virtual scenes replicate actual driving conditions without exposing testers to real physical risks, thus maintaining testing effectiveness while eliminating safety hazards.
Solution Approach 2:
The patent introduces a virtual testing environment as an intermediary between the test subject and real road conditions. This intermediary layer allows indirect testing of SOTIF performance in realistic scenarios without direct exposure to harmful real-world conditions, resolving the contradiction between testing effectiveness and tester safety.
2Reliability
If real road tests are conducted for SOTIF testing, then actual driving conditions are covered, but the testing scope and consistency are limited
Solution Approach 1:
The scene database serves as a universal testing platform that can generate multiple types of driving scenes (weather conditions, road types, traffic situations) within a single system. This multi-functional database enables comprehensive SOTIF testing across diverse conditions without requiring separate real-world test locations, thereby expanding testing scope and consistency.
Solution Approach 2:
The patent segments the testing environment into discrete, controllable scene elements (weather conditions, road types, traffic scenarios) that can be independently selected and combined. This segmentation allows systematic coverage of various driving conditions while maintaining consistent testing protocols, overcoming the limitations of unstructured real road tests.
3Reliability
If comprehensive SOTIF testing is conducted, then more safety issues can be identified, but the testing complexity and time consumption increase
Solution Approach 1:
The patent performs preliminary organization of testing scenes into a structured database with predefined categories and parameters before actual testing begins. This preliminary preparation allows comprehensive SOTIF testing to be conducted systematically without ad-hoc complexity during execution, reducing time consumption while maintaining thorough safety identification.
Solution Approach 2:
The patent manages testing complexity by controlling and varying specific parameters within the virtual environment (weather conditions, traffic density, road types) rather than changing the entire testing system. This parameter-based approach enables comprehensive testing through controlled variations while maintaining system simplicity and reducing overall complexity.
Data Source
AI summary
A comprehensive SOTIF testing system includes a scene database, a virtual testing module and an analyzing module. The scene database includes a plurality of scene groups, and the scene groups respectively correspond to a plurality of source types. Each of the scene groups includes a plurality of scenes, and each of the scenes corresponds to at least one of a plurality of trigger conditions and a plurality of function limitations. The virtual testing module is configured to decide a plurality of to-be-tested members from the source types based on demands of a to-be-tested system, to choose a plurality of chosen members from the scene groups, to generate a plurality of virtual testing scenes based on the scenes of the chosen members, to test the to-be-tested system on the virtual testing scenes, and to confirm whether at least one fail member is present in the virtual testing scenes.